How do you prove that the limit of # (x^3*y^2)/(x^2+y^2)# as (x,y) approaches (0,0) using the epsilon delta proof?

1 Answer
Sep 21, 2017

See explanation

Explanation:

If we have a function #f(x)# of one variable, then:

#lim_(x->c) f(x) = a" "# if and only if:

#AA epsilon > 0 EE delta > 0 : AA x in (c - delta, c + delta)" " abs(f(x) - a) < epsilon#

This can be generalised to a function from #RR^2 -> RR# as follows:

#lim_((x, y) -> (c, d)) f(x, y) = a" "# if and only if:

#AA epsilon > 0 EE delta > 0 : AA x in (c - delta, c + delta) AA y in (d - delta, d + delta)" " abs(f(x, y) - a) < epsilon#

In our example:

#f(x, y) = (x^3y^2)/(x^2+y^2)#

#(c, d) = (0, 0)#

and we will prove that:

#lim_((x, y) -> (0, 0)) = 0#

Given any #epsilon > 0#

Let:

#delta = { (sqrt(2) " if " epsilon > sqrt(2)), (epsilon " if " epsilon <= sqrt(2)) :}#

Let:

#x in (0 - delta, 0 + delta)#

#y in (0 - delta, 0 + delta)#

Then:

#x^2 < delta^2 <= 2#

#y^2 < delta^2 <= 2#

So:

#1/x^2+1/y^2 > 1/2+1/2 = 1#

So:

#1/(1/x^2+1/y^2) < 1#

Then:

If #x = 0# and #y = 0# then #f(x, y)# is undefined. So #(0, 0)# is not part of the domain of #f(x, y)#

If #x = 0# and #y != 0# or #x != 0# and #y = 0# then #f(x, y) = 0# and:

#abs(f(x, y) - 0) = 0 < epsilon#

Otherwise:

#abs(f(x, y) - 0) = abs((x^3y^2)/(x^2+y^2))#

#color(white)(abs(f(x, y) - 0)) = abs(x)(x^2y^2)/(x^2+y^2)#

#color(white)(abs(f(x, y) - 0)) = abs(x)(1)/(1/y^2+1/x^2)#

#color(white)(abs(f(x, y) - 0)) < abs(x) < delta < epsilon#